My station:



Another station across the state:


12 hour difference with odd spike.

Another station closer to mine didn't pick up anything:


Truck driving down the interstate?

I had the same in Warsaw, capital of Poland (Europe), firstly it was in a district Ursynow (c.a. 12 km away), then in Wola, and finally on another bank of Vistula River, the measurements was similar to those shown.

Hi,

These spikes are very interesting; has anyone found out what they were caused by? Your suggestion of a truck is plausible, but some 120 miles with a 12 hour delay could also be a freight train.

Such events seem to require sampling at a higher rate, as both graphs below are single data points. Lukas, were the Warsaw events also single data points?

Is it possible to have the firmware use a 10x sample rate (but reporting to the server only once per hour) and if a selectable value is exceeded, start to report at 10x the rate until it drops back below for 1 hour? Similar to video recording on motion.

--Norbert
Toronto, Canada
(new member, just bought my sensor)

After looking around I now see that the raw data can be obtained using a web query from the individual sensor to get the last 31 days of data (this is very cool - see Number Crunching message board, Message 574), and that older data is also available at http://radioactiveathome.org/boinc/download/data/

But this strange spike from Dec 2013 is not there, because the older data is for 2011, 2012, and the first half of 2013. There is a gap between then and 31 days ago. Is this data still available?

--Norbert (Toronto)

Thanks, Krzysztof for your reply. The sql download script from the forum shows the web query syntax, and I ran a few tests on some sensors. With .../gettrickledata.php?Start=0, the earliest records are seen, all with a date of Nov 15, 2014 (66 days worth). Older data is not available at this url.

The older data that is at http://radioactiveathome.org/boinc/download/data/ is a folder of files all with the same file date. It appears that they were all created in July 2013 (yearly files for 2011, 2012 and 6 monthly files from 2013). It would appear that annual file archives were intended going forward.

At any rate, a similar spike from Bloomingdale appeared on Nov 24, 2014:

http://radioactiveathome.org/scripts/graph/scrollgraph.php?days=54&hostid=12833.
But this one was not accompanied by any spikes from nearby detectors. My curiosity can only be sated if I see a future spike from my own detector, and if I drive around with a portable geiger counter to gauge the size/direction of the event.

--Norbert